The shoulder joint is the most un-constrained joint in the human body and can present numerous clinical challenges including instability, soft tissues injuries and glenohumeral articulation degenerative diseases (1). Studying shoulder biomechanics provides better understanding of shoulder function and the impacts of pathologies and surgical interventions. Measuring joint biomechanics in vivo is challenging as it requires either invasive techniques or radiation imaging such as fluoroscopy. Non-invasive modelling tools have therefore been developed to predict shoulder mechanics. The goal of this study is to use computational modelling, specifically OpenSim, to develop a shoulder model. Concurrent Optimization of Muscles Activation and Kinematics (COMAK) algorithm is used to calculate muscle and joint reaction forces. The calibration and validation of the computational shoulder model will be achieved using data generated from our in-house 8 muscle-actuated, 6 degrees-of-freedom advanced in-vitro cadaveric shoulder simulator. Preliminary OpenSim simulations using passive laxity testing and simple arm motion have been performed with promising results. The development of the calibration model is currently in progress. Once validated, this model will be used to investigate the biomechanics of the shoulder for various clinical challenges along with surgical techniques.